Methods › General › Stochastic Optimization › AdaDelta

AdaDelta

17 papers tagged archive 2025-07-28

Introduced by Matthew D. Zeiler in ADADELTA: An Adaptive Learning Rate Method

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

AdaDelta is a stochastic optimization technique that allows for per-dimension learning rate method for SGD. It is an extension of Adagrad that seeks to reduce its aggressive, monotonically decreasing learning rate. Instead of accumulating all past squared gradients, Adadelta restricts the window of accumulated past gradients to a fixed size w.

Instead of inefficiently storing w previous squared gradients, the sum of gradients is recursively defined as a decaying average of all past squared gradients. The running average E[g²]ₜ at time step t then depends only on the previous average and current gradient:

E[g²]ₜ = γE[g²]ₜ₋₁ + (1-γ)g²ₜ

Usually γ is set to around $0.9$. Rewriting SGD updates in terms of the parameter update vector:

Δθₜ = -η·g_(t, i) θₜ₊₁ = θₜ + Δθₜ

AdaDelta takes the form:

Δθₜ = -η/(√(E[g²]ₜ + ϵ))gₜ

The main advantage of AdaDelta is that we do not need to set a default learning rate.

PaperSourceSee Code · pytorch/pytorch

Papers archive 2025-07-28

17 shown of 17, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

Tasks archive 2025-07-28

11 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Stochastic Optimization6
Image Classification2
Benchmarking1
Cancer Classification1
Federated Learning1
General Classification1
Image Categorization1
Multi-class Classification1
Quantization1
Transfer Learning1
image-classification1

Usage over time archive 2025-07-28

Papers per year tagged with AdaDelta: 2012 to 2024, peak 3 3 0 2012: 1 paper 2012 2013: 0 papers 2013 2014: 0 papers 2014 2015: 1 paper 2015 2016: 0 papers 2016 2017: 0 papers 2017 2018: 1 paper 2018 2019: 3 papers 2019 2020: 0 papers 2020 2021: 3 papers 2021 2022: 3 papers 2022 2023: 3 papers 2023 2024: 2 papers 2024
Papers per year the archive tags with this method, by the paper's archive date (17 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Stochastic Optimization

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections